This research project seeks to comprehensively determine signaling pathway dependencies downstream from oncogenic KRAS in mouse and human pancreatic cancer. While KRAS itself has so far proven refractory to direct pharmacologic targeting, an array of potentially druggable signaling effectors convey signaling downstream from mutant KRAS. These effectors include: 1) the RAF/MEK/ERK kinase cascade; 2) Phospholipid signaling mediated by PLC/PKC and PI3K/AKT; and 3) GTPase signaling mediating by RALA/B and RAC1. A small subset of these downstream effectors have been interrogated for requisite roles in the initiation of pancreatic cancer precursors. However, none have been genetically evaluated with respect to a requisite role in the maintenance of established pancreatic cancer, leaving substantial gaps in our ability to rationally plan therapeutic targeting of these pathways. We therefore propose to genetically dissect these potentially druggable downstream mediators in both mouse and human model systems. The central hypotheses of this project are: First, that the individual components of the composite phenotype induced by mutant Kras will be dependent upon different downstream signaling mediators; second, that combined genetic and pharmacologic targeting of downstream mediators will reveal novel therapeutic vulnerabilities; and third, that different mutant Kras alleles may be differentially sensitive to therapeutic disruption of individual signaling pathways. To test these hypothesis, we propose the following Specific Aims: 1) To create new GEMM ESC-based murine models allowing for the timed inactivation of Raf1, Pik3ca, PLC?, RalA/B and Rac1 in established pancreatic tumors; 2) To combine genetic and pharmacologic inhibition of different downstream KRAS mediators to identify new therapeutic synergies and vulnerabilities using a human pancreatic cancer 3D organoid culture system; and 3) To create and compare new ESC-based genetically engineered mouse models involving Kras G12R and Q61H. The project leverages highly innovative Speedy GEMM ESC mouse technology, including a new dedicated blastocyst injection facility, as well as novel techniques for 3D organoid culture of patient-derived tumor organoids. Together, these studies will for the first tim provide a comprehensive analysis of downstream signaling dependencies in established Kras-driven pancreatic cancer. We anticipate that these studies will inform the design of future clinica trials in which these dependencies are exploited for therapeutic gain.